The Non-Identity Problem from Reasons and Persons by Derek Parfit (1984)

The Non-Identity Problem from Reasons and Persons by Derek Parfit (1984) Each of us might never have existed. What would have made this true? The answer produces a problem that most of us overlook. One of my aims [here] is to discuss this problem. My other aim is to discuss the part of our moral theory in which this problem arises. This is the part that covers how we affect future generations. This is the most important part of our moral theory, since the next few centuries will be the most important in human history. 119. How Our Identity in Fact Depends on When We Were Conceived What would have made it true that some particular person would never have existed? With one qualification, I believe The Time-Dependence Claim: If any particular person had not been conceived when he was in fact conceived, it is in fact true that he would never have existed. This claim is not obviously true. Thus one woman writes: It is always fascinating to speculate on who we would have been if our parents had married other people. In wondering who she would have been, this woman ignores the answer: No one. Though the Time-Dependence Claim is not obviously true, it is not controversial, and it is easy to believe. As I have said, the claim should be qualified. Each of us grew from a particular pair of cells: an ovum and the spermatozoon by which, out of millions, it was fertilized. Suppose that my mother had not conceived a child at the time when in fact she conceived me. And suppose that she had conceived a child within a few days of this time. This child would have grown from the same particular ovum from which I grew. But even if this child had been conceived only a few seconds earlier or later, it is almost certain that he would have grown from a different spermatozoon. This child would have had some but not all of my genes. Would this child have been me? Since I want my Time-Dependence Claim not to be controversial the claim can become (TD2) If any particular person had not been conceived within a month of the time when he was in fact conceived, he would in fact never have existed. I claim that this is in fact true. I do not claim that it is necessarily true. The different views about this subject make competing claims about what is necessary. It is

because I claim less that my claim is not controversial. Those who disagree about what could have happened may agree about what would in fact have happened. As I shall argue, the holders of all plausible views would agree with me. These views make claims about the necessary properties of each particular person. Some of a person's necessary properties are had by everyone: these are the properties that are necessary to being a person. What concerns us here are the distinctive necessary properties of each particular person. Suppose I claim that P is one of Kant's distinctive necessary properties. This means that Kant could not have lacked P, and that only Kant could have had P. According to The Origin View, each person has this distinctive necessary property: that of having grown from the particular pair of cells from which this person in fact grew. This property cannot be fully distinctive. Any pair of identical twins both grew from such a pair of cells. And any fertilized ovum might have later split, and produced twins. The Origin View must be revised to meet this problem. But I need not discuss this revision. It is enough for my purposes that, on this view, Kant could not have grown from a different pair of cells. It is irrelevant that, because there can be twins, it is false that only Kant could have grown from this pair of cells. Holders of the Origin View would accept my claim that, if Kant had not been conceived within a month of the time when he was conceived, he would in fact never have existed. If he had not been conceived in that month, no child would in fact have grown from the particular pair of cells from which he grew. (This claim makes an assumption both about the distinctive necessary properties of this pair of cells, and about the human reproductive system. But these assumptions are not controversial.) On all of the plausible views, my Time-Dependence Claim is true. This claim applies to everyone. You were conceived at a certain time. It is in fact true that, if you had not been conceived within a month of that time, you would never have existed. 120. The Three Kinds of Choice Unless we, or some global disaster, destroy the human race, there will be people living later who do not now exist. These are future people. Science has given to our generation great ability both to affect these people, and to predict these effects. Two kinds of effect raise puzzling questions. We can affect the identities of future people, or who the people are who will later live. And we can affect the number of future people. These effects give us different kinds of choice.

In comparing any two acts, we can ask: Different Number Choices affect both the number and the identities of future people. Same Number Choices affect the identities of future people, but do not affect their number. Same People Choices affect neither. 121. What Weight Should We Give to the Interests of Future People? Most of our moral thinking is about Same People Choices. As I shall argue, such choices are not as numerous as most of us assume. Very many of our choices will in fact have some effect on both the identities and the number of future people. But in most of these cases, because we cannot predict what the particular effects would be, these effects can be morally ignored. We can treat these cases as if they were Same People Choices. In some cases we can predict that some act either may or will be against the interests of future people. This can be true when we are making a Same People Choice. In such a case, whatever we choose, all and only the same people will ever live. Some of these people will be future people. Since these people will exist whatever we choose, we can either harm or benefit these people in a quite straightforward way. Suppose that I leave some broken glass in the undergrowth of a wood. A hundred years later this glass wounds a child. My act harms this child. If I had safely buried the glass, this child would have walked through the wood unharmed. Does it make a moral difference that the child whom I harm does not now exist? On one view, moral principles cover only people who can reciprocate, or harm and benefit each other. If I cannot be harmed or benefited by this child, as we can plausibly suppose, the harm that I cause this child has no moral importance. I assume that we should reject this view.

Some writers claim that, while we ought to be concerned about effects on future people, we are morally justified in being less concerned about effects in the further future. This is a common view in welfare economics, and cost-benefit analysis. On this view, we can discount the more remote effects of our acts and policies, at some rate of n per cent per year. This is called the Social Discount Rate. Suppose we are considering how to dispose safely of the radio-active matter called nuclear waste. If we believe in the Social Discount Rate, we shall be concerned with safety only in the nearer future. We shall not be troubled by the fact that some nuclear waste will be radio-active for thousands of years. At a discount rate of five per cent, one death next year counts for more than a billion deaths in 500 years. On this view, catastrophes in the further future can now be regarded as morally trivial. As this case suggests, the Social Discount Rate is indefensible. Remoteness in time roughly correlates with some important facts, such as predictability. But, as I argue in Appendix F, these correlations are too rough to justify the Social Discount Rate. The present moral importance of future events does not decline at a rate of n per cent per year. Remoteness in time has, in itself, no more significance than remoteness in space. Suppose that I shoot some arrow into a distant wood, where it wounds some person. If I should have known that there might be someone in this wood, I am guilty of gross negligence. Because this person is far away, I cannot identify the person whom I harm. But this is no excuse. Nor is it any excuse that this person is far away. We should make the same claims about effects on people who are temporally remote. 122. A Young Girl's Child Future people are, in one respect, unlike distant people. We can affect their identity. And many of our acts have this effect. This fact produces a problem. Before I describe this problem, I shall repeat some preliminary remarks. I assume that one person can be worse off than another, in morally significant ways, and by more or less. But I do not assume that these comparisons could be, even in principle, precise. I assume that there is only rough or partial comparability. On this assumption, it could be true of two people that neither is worse off than the other, but this would not imply that these people are exactly equally well off. Worse off could be taken to refer, either to someone's level of happiness, or more narrowly to his standard of living, or, more broadly, to the quality of his life. Since it is the broadest, I shall often use the phrase the quality of life. I also call certain lives worth living. This description can be ignored by those who believe that there could not be lives that are not worth living. But, like many other people, I believe that there could be such lives. Finally, I extend the ordinary use of the phrase worth living. If one of two people would have a lower quality of life, I call his life to this extent less worth living.

When considering future people, we must answer two questions: (1) If we cause someone to exist, who will have a life worth living, do we thereby benefit this person? (2) Do we also benefit this person if some act of ours is a remote but necessary part of the cause of his existence? These are difficult questions. If we answer Yes to both, I shall say that we believe that causing to exist can benefit. Some people answer Yes to (1) but No to (2). These people give their second answer because they use benefit in its ordinary sense. As I argued in Section 25, we ought for moral purposes to extend our use of benefit. If we answer Yes to (1) we should answer Yes to (2). Many people answer No to both these questions. These people might say: We benefit someone if it is true that, if we had not done what we did, this would have been worse for this person. If we had not caused someone to exist, this would not have been worse for this person. I believe that, while it is defensible to answer No to both these questions, it is also defensible to answer Yes to both. For those who doubt this second belief I have written Appendix G. Since I believe that it is defensible both to claim and to deny that causing to exist can benefit, I shall discuss the implications of both views. Consider The 14-Year-Old Girl. This girl chooses to have a child. Because she is so young, she gives her child a bad start in life. Though this will have bad effects throughout this child's life, his life will, predictably, be worth living. If this girl had waited for several years, she would have had a different child, to whom she would have given a better start in life. Since such cases are becoming common, they raise a practical problem. 8 They also raise a theoretical problem. Suppose that we tried to persuade this girl that she ought to wait. We claimed: If you have a child now, you will soon regret this. If you wait, this will be better for you. She replied: This is my affair. Even if I am doing what will be worse for me, I have a right to do what I want. We replied: This is not entirely your affair. You should think not only of yourself, but also of your child. It will be worse for him if you have him now. If you have him later, you will give him a better start in life. We failed to persuade this girl. She had a child when she was 14, and, as we predicted, she gave him a bad start in life. Were we right to claim that her decision

was worse for her child? If she had waited, this particular child would never have existed. And, despite its bad start, his life is worth living. Suppose first that we do not believe that causing to exist can benefit. We should ask, If someone lives a life that is worth living, is this worse for this person than if he had never existed? Our answer must be No. Suppose next that we believe that causing to exist can benefit. On this view, this girl's decision benefits her child. On both views, this girl's decision was not worse for her child. When we see this, do we change our mind about this decision? Do we cease to believe that it would have been better if this girl had waited, so that she could give to her first child a better start in life? I continue to have this belief, as do most of those who consider this case. But we cannot defend this belief in the natural way that I suggested. We cannot claim that this girl's decision was worse for her child. What is the objection to her decision? This question arises because, in the different outcomes, different people would be born. I shall therefore call this the Non-Identity Problem. It may be said: In one sense, this girl's decision was worse for her child. In trying to persuade this girl not to have a child now, we can use the phrase her child and the pronoun he to cover any child that she might have. These words need not refer to one particular child. We can truly claim: If this girl does not have her child now, but waits and has him later, he will not be the same particular child. If she has him later, he will be a different child. By using these words in this way, we can explain why it would be better if this girl waits. We can claim: (A) The objection to this girl's decision is that it will probably be worse for her child. If she waited, she would probably give him a better start in life. Though we can truly make this claim, it does not explain the objection to this girl's decision. This becomes clear after she has had her child. The phrase her child now naturally refers to this particular child. And this girl's decision was not worse for this child. Though there is a sense in which (A) is true, (A) does not appeal to a familiar moral principle. On one of our familiar principles, it is an objection to someone's choice that this choice will be worse for, or be against the interests of, any other particular person. If we claim that this girl's decision was worse for her child, we cannot be claiming that it was worse for a particular person. We cannot claim, of the girl's child, that her decision was worse for him. We must admit that, in claim (A), the words her child do not refer to her child. (A) is not about what is good or bad for any of the particular people who ever live. (A) appeals to a new principle, that must be explained and justified. If (A) seems to appeal to a familiar principle, this is because it has two senses. Here is another example. A general shows military skill if, in many battles, he always

makes his the winning side. But there are two ways of doing this. He might win victories. Or he might always, when he is about to lose, change sides. A general shows no military skill if it is only in the second sense that he always makes his the winning side. To what principle does (A) appeal? We should state the principle in a way that shows the kind of choice to which it applies. These are Same Number Choices, which affect the identities of future people, but do not affect their number. We might suggest The Same Number Quality Claim, or Q: If in either of two possible outcomes the same number of people would ever live, it would be worse if those who live are worse off, or have a lower quality of life, than those who would have lived. This claim is plausible. And it implies what we believe about the 14-Year-Old Girl. The child that she has now will probably be worse off than a child she could have had later would have been, since this other child would have had a better start in life. If this is true, Q implies that this is the worse of these two outcomes. Q implies that it would have been better if this girl had waited, and had a child later. We may shrink from claiming, of this girl's actual child, that it would have been better if he had never existed. But, if we claimed earlier that it would be better if this girl waits, this is what we must claim. We cannot consistently make a claim and deny this same claim later. If (1) in 1990 it would be better if this girl waits and has a child later, then (2) in 2020 it would have been better if she had waited and had a child later. And (2) implies (3) that it would have been better if the child who existed had not been her actual child. If we cannot accept (3), we must reject (1). I suggest that, on reflection, we can accept (3). I believe that, if I was the actual child of this girl, I could accept (3). (3) does not imply that my existence is bad, or intrinsically morally undesirable. The claim is merely that, since a child born later would probably have had a better life than mine, it would have been better if my mother had waited, and had a child later. This claim need not imply that I ought rationally to regret that my mother had me, or that she ought rationally to regret this. Since it would have been better if she had waited, she ought perhaps to have some moral regret. And it is probably true that she made the outcome worse for herself. But, even if this is true, it does not show that she ought rationally to regret her act, all things considered. If she loves me, her actual child, this is enough to block the claim that she is irrational if she does not have such regret. Even when it implies a claim like (3), I conclude that we can accept Q. Though Q is plausible, it does not solve the Non-Identity Problem. Q covers only the cases where, in the different outcomes, the same number of people would ever live. We need a claim that covers cases where, in the different outcomes, different numbers would ever live. The Non-Identity Problem can arise in these cases.

Because Q is restricted, it could be justified in several different ways. There are several principles that imply Q, but conflict when applied to Different Number Choices. We shall need to decide which of these principles, or which set of principles, we ought to accept. Call what we ought to accept Theory X. X will solve the Non-Identity Problem in Different Number Choices. And X will tell us how Q should be justified, or more fully explained. In the case of the 14-Year-Old Girl, we are not forced to appeal to Q. There are other facts to which we could appeal, such as the effects on other people. But the problem can arise in a purer form. 123. How Lowering the Quality of Life Might Be Worse for No One Suppose that we are choosing between two social or economic policies. And suppose that, on one of the two policies, the standard of living would be slightly higher over the next century. This effect implies another. It is not true that, whichever policy we choose, the same particular people will exist in the further future. Given the effects of two such policies on the details of our lives, it would increasingly over time be true that, on the different policies, people married different people. And, even in the same marriages, the children would increasingly over time be conceived at different times. As I have argued, children conceived more than a month earlier or later would in fact be different children. Since the choice between our two policies would affect the timing of later conceptions, some of the people who are later born would owe their existence to our choice of one of the two policies. If we had chosen the other policy, these particular people would never have existed. And the proportion of those later born who owe their existence to our choice would, like ripples in a pool, steadily grow. We can plausibly assume that, after one or two centuries, there would be no one living in our community who would have been born whichever policy we chose. (It may help to think about this question: how many of us could truly claim, Even if railways and motor cars had never been invented, I would still have been born?) How does this produce a problem? Consider Depletion. As a community, we must choose whether to deplete or conserve certain kinds of resources. If we choose Depletion, the quality of life over the next two centuries would be slightly higher than it would have been if we had chosen Conservation. But it would later, for many centuries, be much lower than it would have been if we had chosen Conservation. This would be because, at the start of this period, people would have to find alternatives for the resources that we had depleted. It is worth distinguishing two versions of this case. The effects of the different policies would be as shown below.

We could never know, in such detail, that these would be the effects of two policies. But this is no objection to this case. Similar effects would sometimes be predictable. Nor does it matter that this imagined case is artificially simple, since this merely clarifies the relevant questions. Suppose that we choose Depletion, and that this has either of the two effects shown in my diagram. Is our choice worse for anyone? Because we chose Depletion, millions of people have, for several centuries, a much lower quality of life. This quality of life is much lower, not than it is now, but than it would have been if we had chosen Conservation. These people's lives are worth living; and, if we had chosen Conservation, these particular people would never have existed. Suppose that we do not assume that causing to exist can benefit. We should ask, If particular people live lives that are worth living, is this worse for these people than if they had never existed? Our answer must be No. Suppose next that we do assume that causing to exist can benefit. Since these future people's lives will be worth living, and they would never have existed if we had chosen Conservation, our choice of Depletion is not only not worse for these people: it benefits them. On both answers, our choice will not be worse for these future people. Moreover, when we understand the case, we know that this is true. We know that, even if it greatly lowers the quality of life for several centuries, our choice will not be worse for anyone who ever lives.

Does this make a moral difference? There are three views. It might make all the difference, or some difference, or no difference. There might be no objection to our choice, or some objection, or the objection may be just as strong. Some believe that what is bad must be bad for someone. On this view, there is no objection to our choice. Since it will be bad for no one, our choice cannot have a bad effect. The great lowering of the quality of life provides no moral reason not to choose Depletion. Certain writers accept this conclusion. But it is very implausible. Before we consider cases of this kind, we may accept the view that what is bad must be bad for someone. But the case of Depletion shows, I believe, that we must reject this view. The great lowering of the quality of life must provide some moral reason not to choose Depletion. This is believed by most of those who consider cases of this kind. If this is what we believe, we should ask two questions: (1) What is the moral reason not to choose Depletion? (2) Does it make a moral difference that this lowering of the quality of life will be worse for no one? Would this effect be worse, having greater moral weight, if it was worse for particular people? Our need to answer (1), and other similar questions, I call the Non-Identity Problem. This problem arises because the identities of people in the further future can be very easily affected. Some people believe that this problem is a mere quibble. This reaction is unjustified. The problem arises because of superficial facts about our reproductive system. But, though it arises in a superficial way, it is a real problem. When we are choosing between two social or economic policies, of the kind that I described, it is not true that, in the further future, the same people will exist whatever we choose. It is therefore not true that a choice like Depletion will be against the interests of future people. We cannot dismiss this problem with the pretense that this is true. We partly answer question (1) if we appeal to Q. On this claim, if the numbers would be the same, it would be worse if those who live have a lower quality of life than those who would have lived. But the problem can arise in cases where, in the different outcomes, there would be different numbers of people. To cover these cases we need Theory X. Only X will explain how Q should be justified, and provide a full solution to our problem. 124. Why an Appeal to Rights Cannot Wholly Solve the Problem Can we solve our problem by appealing to people's rights? Reconsider the 14-Year- Old Girl. By having her child so young, she gives him a bad start in life. It might

be claimed: The objection to this girl's decision is that she violates her child's right to a good start in life. Even if this child has this right, it could not have been fulfilled. This girl could not have had this child when she was a mature woman. Some would claim that, since this child's right could not be fulfilled, this girl cannot be claimed to violate his right. The objector might reply: It is wrong to cause someone to exist if we know that this person will have a right that cannot be fulfilled. Can this be the objection to this girl's decision? Some years ago, a British politician welcomed the fact that, in the previous year, there had been fewer teenage pregnancies. A middle-aged man wrote in anger to The Times. He had been born when his mother was only 14. He admitted that, because his mother was so young, his early years had been hard for both of them. But his life was now well worth living. Was the politician suggesting that it would have been better if he had never been born? This suggestion seemed to him outrageous. The politician was, implicitly, suggesting this. On the politician's view, it would have been better if this man's mother had waited for several years before having children. I believe that we should accept this view. But can we plausibly explain this view by claiming that this angry man had a right that was not fulfilled? I believe that we cannot. Suppose that I have a right to privacy. I ask you to marry me. If you accept, you are not acting wrongly, by violating my right to privacy. Since I am glad that you act as you do, with respect to you I waive this right. A similar claim applies to the writer of the angry letter to The Times. On the suggestion made above, this man has a right to be born by a mature woman, who would give him a good start in life. This man's mother acted wrongly because she caused him to exist with a right that cannot be fulfilled. But this man's letter shows that he was glad to be alive. He denies that his mother acted wrongly because of what she did to him. If we had claimed that her act was wrong, because he has a right that cannot be fulfilled, he could have said, I waive this right. This would have undermined our objection to his mother's act. It would have been better if this man's mother had waited. But this is not because of what she did to her actual child. It is because of what she could have done for any child that she could have had when she was mature. The objection must be that, if she had waited, she could have given to some other child a better start in life. Return now to the Case of Depletion. Suppose that we choose Greater Depletion. More than two centuries later, the quality of life is much lower than it would have been if we had chosen Conservation. But the people who will then be living will have a quality of life that is about as high as ours will on average be over the next century. Do these people have rights to which an objector can appeal?

It might be claimed that these people have a right to their share of the resources that we have depleted. But people do not have rights to a share of a particular resource. Suppose that we deplete some resource, but invent technology that will enable our successors, though they lack this resource, to have the same range of opportunities. There would be no objection to what we have done. The most that could be claimed is that people in each generation have a right to an equal range of opportunities, or to an equally high quality of life. If we choose Greater Depletion, those who live more than two centuries later will have fewer opportunities, and a lower quality of life, than some earlier and some later generations. If people have a right to equal opportunities, and an equally high quality of life, an appeal to these rights may provide some objection to our choice. Those who live more than two centuries later could not possibly have had greater opportunities, or a higher quality of life. If we had chosen otherwise, these people would never have existed. Since their rights could not be fulfilled, we may not violate their rights. But, as before, it may be objected that we cause people to exist with rights that cannot be fulfilled. It is not clear that this is a good objection. If these people knew the facts, they would not regret that we acted as we did. If they were glad to be alive, they might react like the man who wrote to The Times. They might waive their rights. But, since we cannot assume that this is how they would all react, an appeal to their rights may provide some objection to our choice. Can this appeal provide an objection to our choice of Lesser Depletion? In this case, those who live more than two centuries later have a much higher quality of life than we do now. Can we claim that these people have a right to an even higher quality of life? I believe that, on any plausible theory about rights, the answer would be No. It will help to imagine away the Non-Identity Problem. Suppose that our reproductive system was very different. Suppose that, whatever policies we followed, the very same people would live more than two centuries later. The objection to our choice would then be that, for the sake of small benefits to ourselves and our children, we prevent many future people from receiving very much greater benefits. Since these future people would be better off than us, we would not be acting unjustly. The objection to our choice would have to appeal to the Principle of Utility. Could this objection appeal to rights? Only if, like Godwin, we present Utilitarianism as a theory about rights. On Godwin's view, everyone has a right to get what the Principle of Utility implies that he should be given. Most of those who believe in rights would reject this view. Many people explain rights as what constrain, or limit, the Principle of Utility. These people claim that it is wrong to violate certain rights, even if this would greatly increase the net sum of benefits minus burdens. On such a theory, some weight is given to the Principle of Utility.

Since such a theory is not Utilitarian, this principle is better called the Principle of Beneficence. This principle is one part of such a theory, and the claim that we have certain rights is a different part of this theory. I shall assume that, if we believe in rights, this is the kind of moral theory that we accept. Return to the case where we imagine away the Non-Identity Problem. If we reject Godwin's view, we could not object to the choice of Lesser Depletion by appealing to the rights of those who will live in the further future. Our objection would appeal to the Principle of Beneficence. The objection would be that, for the sake of small benefits to ourselves and our children, we deny, to people better off than us, very much greater benefits. In calling this an objection, I need not claim that it shows our choice to be wrong. I am merely claiming that, since we deny these people very much greater benefits, this provides some moral reason not to make this choice. If we now restore our actual reproductive system, this reason disappears. Consider the people who will live more than two centuries later. Our choice of Lesser Depletion does not deny these people any benefit. If we had chosen Conservation, this would not have benefited these people, since they would never have existed. When we assume away the Non-Identity Problem, our reason not to make this choice is explained by an appeal, not to people's rights, but to the Principle of Beneficence. When we restore the Non-Identity Problem, this reason disappears. Since this reason appealed to the Principle of Beneficence, what the problem shows is that this principle is inadequate, and must be revised. We need a better account of beneficence, or what I call Theory X. One part of our moral theory appeals to beneficence; another part appeals to people's rights. We should therefore not expect that an appeal to rights could fill the gap in our inadequate Principle of Beneficence. We should expect that, as I have claimed, appealing to rights cannot wholly solve the Non-Identity Problem. 125. Does the Fact of Non-Identity Make a Moral Difference? In trying to revise our Principle of Beneficence trying to find Theory X we must consider cases where, in the different outcomes, different numbers of people would exist. Before we turn to these cases, we can ask what we believe about the other question that I mentioned. Our choice of Depletion will be worse for no one. Does this make a moral difference? We may be able to remember a time when we were concerned about effects on future generations, but had overlooked the Non-Identity Problem. We may have thought that a policy like Depletion would be against the interests of future people. When we saw that this was false, did we become less concerned about effects on future generations?

When I saw the problem, I did not become less concerned. And the same is true of many other people. I shall say that we accept the No-Difference View. It is worth considering a different example: The Medical Programs. There are two rare conditions, J and K, which cannot be detected without special tests. If a pregnant woman has Condition J, this will cause the child she is carrying to have a certain handicap. A simple treatment would prevent this effect. If a woman has Condition K when she conceives a child, this will cause this child to have the same particular handicap. Condition K cannot be treated, but always disappears within two months. Suppose next that we have planned two medical programs, but there are funds for only one; so one must be cancelled. In the first program, millions of women would be tested during pregnancy. Those found to have Condition J would be treated. In the second program, millions of women would be tested when they intend to try to become pregnant. Those found to have Condition K would be warned to postpone conception for at least two months, after which this incurable condition will have disappeared. Suppose finally that we can predict that these two programs would achieve results in as many cases. If there is Pregnancy Testing, 1,000 children a year will be born normal rather than handicapped. If there is Preconception Testing, there will each year be born 1,000 normal children rather than a 1,000, different, handicapped children. Would these two programs be equally worthwhile? Let us note carefully what the difference is. As a result of either program, 1,000 couples a year would have a normal rather than a handicapped child. These would be different couples, on the two programs. But since the numbers would be the same, the effects on the parents and on other people would be morally equivalent. If there is a moral difference, this can only be in the effects on the children. Note next that, in judging these effects, we need have no view about the moral status of a fetus. We can suppose that it would take a year before either kind of testing could begin. When we choose between the two programs, none of the children has yet been conceived. And all those who are conceived will become adults. We are therefore considering effects, not on present fetuses, but on future people. Assume next that the handicap in question, though it is not trivial, is not so severe as to make life doubtfully worth living. Even if it can be against our interests to have been born, this is not true of those born with this handicap. Since we cannot afford both programs, which should we cancel? Under one description, both would have the same effect. Suppose that Conditions J and K are the only causes of this handicap. The incidence is now 2,000 among those born in each year. Either program would halve the incidence; the rate would drop to 1,000 a year. The difference is this. If we decide to cancel Pregnancy Testing, it will be true of those who are later born handicapped that, but for our decision, they would

have been cured. Our decision will be worse for all these people. If instead we decide to cancel Pre-Conception Testing, there will later be just as many people who are born with this handicap. But it would not be true of these people that, but for our decision, they would have been cured. These people owe their existence to our decision. If we had not decided to cancel Pre-Conception Testing, the parents of these handicapped children would not have had them. They would have later had different children. Since the lives of these handicapped children are worth living, our decision will not be worse for any of them. Does this make a moral difference? Or are the two programs equally worthwhile? Is all that matters morally how many future lives will be lived by normal rather than handicapped people? Or does it also matter whether these lives would be lived by the very same people? We should add one detail to the case. If we decide to cancel Pregnancy Testing, those who are later born handicapped might know that, if we had made a different decision, they would have been cured. Such knowledge might make their handicap harder to bear. We should therefore assume that, though it is not deliberately concealed, these people would not know this fact. With this detail added, I judge the two programs to be equally worthwhile. I know of some people who do not accept this claim; but I know of more who do. My reaction is not merely an intuition. It is the judgement that I reach by reasoning as follows. Whichever program is cancelled, there will later be just as many people with this handicap. These people would be different in the two outcomes that depend on our decision. And there is a claim that applies to only one of these two groups of handicapped people. Though they do not know this fact, the people in one group could have been cured. I therefore ask: If there will be people with some handicap, the fact that they are handicapped is bad. Would it be worse if, unknown to them, their handicap could have been cured? This would be worse if this fact made these people worse off than people whose handicap could not have been cured. But this fact does not have this effect. If we decide to cancel Pregnancy Testing, there will be a group of handicapped people. If we decide to cancel Pre- Conception Testing, there will be a different group of handicapped people. The people in the first group would not be worse off than the people in the second group would have been. Since this is so, I judge these two outcomes to be morally equivalent. Given the details of the case, it seems to me irrelevant that one of the groups but not the other could have been cured. This fact would have been relevant if curing this group would have reduced the incidence of this handicap. But, since we have funds for only one program, this is not true. If we choose to cure the first group, there will later be just as many people with this handicap. Since curing the first group would not reduce the number who will be handicapped, we ought to choose to cure this group only if they have a stronger claim to be cured. And they do not have a stronger claim. If we could cure

the second group, they would have an equal claim to be cured. If we chose to cure the first group, they would merely be luckier than the second group. Since they would merely be luckier, and they do not have a stronger claim to be cured, I do not believe that we ought to choose to cure these people. Since it is also true that, if we choose to cure these people, this will not reduce the number of people who will be handicapped, I conclude that the two programs are equally worthwhile. If Pre-Conception Testing would achieve results in a few more cases, I would judge it to be the better program. This matches my reaction to our choice of Depletion. I believe that it would be bad if there would later be a great lowering of the quality of life. And I believe that it would not be worse if the people who later live would themselves have existed if we had chosen Conservation. The bad effect would not be worse if it had been, in this way, worse for any particular people. In considering both cases, I accept the No-Difference View. So do many other people. 127. Conclusions I shall now summarize what I have claimed. It is in fact true of everyone that, if he had not been conceived within a month of the time when he was conceived, he would never have existed. Because this is true, we can easily affect the identities of future people, or who the people are who will later live. If a choice between two social policies will affect the standard of living or the quality of life for about a century, it will affect the details of all the lives that, in our community, are later lived. As a result, some of those who later live will owe their existence to our choice of one of these two policies. After one or two centuries, this will be true of everyone in our community. This fact produces a problem. One of these two policies may, in the further future, cause a great lowering of the quality of life. This would be the effect of the policy I call Depletion. This effect is bad, and provides a moral reason not to choose Depletion. But, because of the fact just mentioned, our choice of Depletion will be worse for no one. Some people believe that a choice cannot have bad effects if this choice will be worse for no one. The Case of Depletion shows that we must reject this view. Since [Depletion] will be worse for no one, we need to explain why we have a moral reason not to [choose Depletion]. This problem arises because, in the different outcomes, different people would exist. I therefore call this the Non- Identity Problem. I asked whether we can solve this problem by appealing to people's rights. I argued that the objection to our choice cannot appeal only to people's rights. The objection must in part appeal to a claim like Q, which compares different possible lives. And we cannot plausibly appeal to rights in explaining the objection to our choice of Lesser Depletion. Even after the great lowering of the quality of life, those

who will be living will be much better off than we are now. These people cannot be claimed to have a right to the even higher quality of life that different people would have enjoyed if we had chosen Conservation. If we imagine away the Non- Identity Problem, the objection to our choice would appeal to our Principle of Beneficence. To solve the Non-Identity Problem, we must revise this principle. One revised principle is Q, the Same Number Quality Claim. According to Q, if in either of two outcomes there would be the same number of people, it would be worse if those who live are worse off, or have a lower quality of life, than those who would have lived. We need a wider principle to cover cases where, in the different outcomes, there would be different numbers of people. This needed principle I call Theory X. Only X will fully solve the Non-Identity Problem. Does the fact of non-identity make a moral difference? When we see that our choice of Depletion will be worse for no one, we may believe that there is less objection to our choice. But I believe that the objection is just as strong. And I have a similar belief when I compare the effects of the two Medical Programs. This belief I call the No Difference View.